PM: Do not hold dpm_list_mtx while disabling/enabling nonboot CPUs

We shouldn't hold dpm_list_mtx while executing
[disable|enable]_nonboot_cpus(), because theoretically this may lead
to a deadlock as shown by the following example (provided by Johannes
Berg):

CPU 3       CPU 2                     CPU 1
                                      suspend/hibernate
            something:
            rtnl_lock()               device_pm_lock()
                                       -> mutex_lock(&dpm_list_mtx)

            mutex_lock(&dpm_list_mtx)

linkwatch_work
 -> rtnl_lock()
                                      disable_nonboot_cpus()
                                       -> flush CPU 3 workqueue

Fortunately, device drivers are supposed to stop any activities that
might lead to the registration of new device objects way before
disable_nonboot_cpus() is called, so it shouldn't be necessary to
hold dpm_list_mtx over the entire late part of device suspend and
early part of device resume.

Thus, during the late suspend and the early resume of devices acquire
dpm_list_mtx only when dpm_list is going to be traversed and release
it right after that.

This patch is reported to fix the regressions tracked as
http://bugzilla.kernel.org/show_bug.cgi?id=13245.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Reported-by: Miles Lane <miles.lane@gmail.com>
Tested-by: Ming Lei <tom.leiming@gmail.com>

4 files changed, 8 insertions, 26 deletions

diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 69b4ddb..3e4bc69 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -357,6 +357,7 @@ static void dpm_power_up(pm_message_t state)
 {
 	struct device *dev;
 
+	mutex_lock(&dpm_list_mtx);
 	list_for_each_entry(dev, &dpm_list, power.entry)
 		if (dev->power.status > DPM_OFF) {
 			int error;
@@ -366,6 +367,7 @@ static void dpm_power_up(pm_message_t state)
 			if (error)
 				pm_dev_err(dev, state, " early", error);
 		}
+	mutex_unlock(&dpm_list_mtx);
 }
 
 /**
@@ -614,6 +616,7 @@ int device_power_down(pm_message_t state)
 	int error = 0;
 
 	suspend_device_irqs();
+	mutex_lock(&dpm_list_mtx);
 	list_for_each_entry_reverse(dev, &dpm_list, power.entry) {
 		error = suspend_device_noirq(dev, state);
 		if (error) {
@@ -622,6 +625,7 @@ int device_power_down(pm_message_t state)
 		}
 		dev->power.status = DPM_OFF_IRQ;
 	}
+	mutex_unlock(&dpm_list_mtx);
 	if (error)
 		device_power_up(resume_event(state));
 	return error;
diff --git a/kernel/kexec.c b/kernel/kexec.c
index 5a758c6e..e498377 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -1451,7 +1451,6 @@ int kernel_kexec(void)
 		error = device_suspend(PMSG_FREEZE);
 		if (error)
 			goto Resume_console;
-		device_pm_lock();
 		/* At this point, device_suspend() has been called,
 		 * but *not* device_power_down(). We *must*
 		 * device_power_down() now.  Otherwise, drivers for
@@ -1489,7 +1488,6 @@ int kernel_kexec(void)
 		enable_nonboot_cpus();
 		device_power_up(PMSG_RESTORE);
  Resume_devices:
-		device_pm_unlock();
 		device_resume(PMSG_RESTORE);
  Resume_console:
 		resume_console();
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index b0dc9e7..5cb080e 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -215,8 +215,6 @@ static int create_image(int platform_mode)
 	if (error)
 		return error;
 
-	device_pm_lock();
-
 	/* At this point, device_suspend() has been called, but *not*
 	 * device_power_down(). We *must* call device_power_down() now.
 	 * Otherwise, drivers for some devices (e.g. interrupt controllers)
@@ -227,7 +225,7 @@ static int create_image(int platform_mode)
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to power down, "
 			"aborting hibernation\n");
-		goto Unlock;
+		return error;
 	}
 
 	error = platform_pre_snapshot(platform_mode);
@@ -280,9 +278,6 @@ static int create_image(int platform_mode)
 	device_power_up(in_suspend ?
 		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
 
- Unlock:
-	device_pm_unlock();
-
 	return error;
 }
 
@@ -344,13 +339,11 @@ static int resume_target_kernel(bool platform_mode)
 {
 	int error;
 
-	device_pm_lock();
-
 	error = device_power_down(PMSG_QUIESCE);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to power down, "
 			"aborting resume\n");
-		goto Unlock;
+		return error;
 	}
 
 	error = platform_pre_restore(platform_mode);
@@ -403,9 +396,6 @@ static int resume_target_kernel(bool platform_mode)
 
 	device_power_up(PMSG_RECOVER);
 
- Unlock:
-	device_pm_unlock();
-
 	return error;
 }
 
@@ -464,11 +454,9 @@ int hibernation_platform_enter(void)
 		goto Resume_devices;
 	}
 
-	device_pm_lock();
-
 	error = device_power_down(PMSG_HIBERNATE);
 	if (error)
-		goto Unlock;
+		goto Resume_devices;
 
 	error = hibernation_ops->prepare();
 	if (error)
@@ -493,9 +481,6 @@ int hibernation_platform_enter(void)
 
 	device_power_up(PMSG_RESTORE);
 
- Unlock:
-	device_pm_unlock();
-
  Resume_devices:
 	entering_platform_hibernation = false;
 	device_resume(PMSG_RESTORE);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index f99ed6a..8680282 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -289,12 +289,10 @@ static int suspend_enter(suspend_state_t state)
 {
 	int error;
 
-	device_pm_lock();
-
 	if (suspend_ops->prepare) {
 		error = suspend_ops->prepare();
 		if (error)
-			goto Done;
+			return error;
 	}
 
 	error = device_power_down(PMSG_SUSPEND);
@@ -343,9 +341,6 @@ static int suspend_enter(suspend_state_t state)
 	if (suspend_ops->finish)
 		suspend_ops->finish();
 
- Done:
-	device_pm_unlock();
-
 	return error;
 }